Abstract: Computer-based method and system for predicting Land Use Land Cover (LULC) classification of a geographic area using trained deep learning model are described herein. The method and system facilitate a user, without any hard knowledge of geographic information system (GIS) to get LULC classification of a geographic area with a fully automated, single step, and single input process.

Abstract: Computer-based method and system for geo-spatial analysis are disclosed herein. The method and system facilitate a user, without any hard knowledge of geographic information system (GIS) to perform quick geo-spatial analysis with a fully automated, single step, and single input process including automatically retrieving a first set of satellite images corresponding to a geographic area and a first time frame provided by the user, automatically processing the first set of satellite images to determine first values of one or more urban parameters corresponding to the geographic area for the first time frame, and automatically presenting a visualization depicting the first values of the one or more urban parameters.

Abstract: The present invention discloses a single package additive for improving lubricity and conductivity properties of ultra-low sulfur diesel fuels. The single package additive is a reaction product of a fatty acid composition, a glycerol tricarboxylates, a polysulfone, a polyamine, an alkylated benzene sulfonic acid and a phenol derivative. More specifically, the present invention discloses a reaction product of: a fatty acid composition in the range of 60-95% wt/wt; a glycerol tricarboxylate in the range of 0.1-10.0% wt/wt; a polysulfone in the range of 0.1-5.0% wt/wt; a polyamine in the range of 0.1-5.0% wt/wt; an alkylated benzene sulfonic acid in the range of 0.1-5.0% wt/wt; and a phenol derivative in the range of 0.1-10.0% wt/wt. The present invention also discloses a single-pot process for the preparation of said reaction product.

Abstract: Computer-based method and system for geo-spatial analysis for determining groundwater potential zones within a geographic area are disclosed herein. The method and system facilitate a user, without any hard knowledge of geographic information system (GIS) to perform quick geo-spatial analysis with a fully automated, single step, and single input process including automatically retrieving a set of satellite images and digital elevation model (DEM) corresponding to a geographic area and a time frame provided by the user, automatically processing the set of satellite images and DEM to determine one or more spectral indices and topographic wetness index (TWI) corresponding to the geographic area and time frame, automatically determining the groundwater potential zones using the one or more spectral indices and the TWI and automatically presenting a visualization depicting the groundwater potential zones within the geographic area.

Abstract: The present invention relates to preparation of catalyst for production of olefinic hydrocarbons by dehydrogenation of their corresponding paraffins, particularly propylene from propane, comprising a metal oxide or combination of metal oxides utilizing spent catalyst from Fluid Catalytic Cracking (FCC)/Resid Fluid Catalytic Cracking (RFCC) processes. The metal oxides are possibly from transition metal group, particularly from groups VB, VIB, VIII, and Lanthanide series, and at least one metal from alkali group. The catalyst support used is spent catalyst or modified spent catalyst or combination thereof. The said catalyst can be used for both non-oxidative Propane Dehydrogenation (PDH) and Oxidative Propane Dehydrogenation (OPDH) process in the presence of CO2.

Abstract: A process for upgrading light cycle oil (LCO) to a high octane gasoline and a high aromatic feedstock for aromatic complex, by utilizing fluid catalytic cracking (FCC) is disclosed. FCC cracks unconverted diesel range stream from a hydrocracker to a high octane gasoline. This process further improves the quality (e.g., research octane number (RON), and Sulfur) of gasoline by utilizing FCC and hydrocracking process.

Abstract: Comparing extracted card data from a continuous scan comprises receiving, by one or more computing devices, a digital scan of a card; obtaining a plurality of images of the card from the digital scan of the physical card; performing an optical character recognition algorithm on each of the plurality of images; comparing results of the application of the optical character recognition algorithm for each of the plurality of images; determining if a configured threshold of the results for each of the plurality of images match each other; and verifying the results when the results for each of the plurality of images match each other. Threshold confidence level for the extracted card data can be employed to determine the accuracy of the extraction. Data is further extracted from blended images and three-dimensional models of the card. Embossed text and holograms in the images may be used to prevent fraud.

Abstract: The invention discloses a novel method for preparation of highly active and selective dehydrogenation catalyst, comprising of metal oxide of group VIB elements of periodic table, and at least one metal oxide from group IA and/or group VIII, supported on alumina or silica or mixture thereof, wherein the accessibility to active sites and dispersion of metal oxides is enhanced by the addition of carbonaceous material such as coke derived from coal or petroleum coke or any other form of carbon, during catalyst preparation and its combustion thereof during calcination.

Abstract: A computing system and method can be used to implement a version of federated learning (FL) that incorporates adaptivity (e.g., leverages an adaptive learning rate). In particular, the present disclosure provides a general optimization framework in which (1) clients perform multiple epochs of training using a client optimizer to minimize loss on their local data and (2) a server system updates its global model by applying a gradient-based server optimizer to the average of the clients' model updates. This framework can seamlessly incorporate adaptivity by using adaptive optimizers as client and/or server optimizers. Building upon this general framework, the present disclosure also provides example specific adaptive optimization techniques for FL which use per-coordinate methods as server optimizers. By focusing on adaptive server optimization, the use of adaptive learning rates is enabled without increase in client storage or communication costs and compatibility with cross-device FL can be ensured.

Abstract: A process of conversion of fuel grade coke produced through thermal cracking of heavy petroleum residue to anode grade coke. The process employs high sulfur fuel grade coke as the feedstock to produce low sulfur coke, which can be used to manufacture electrodes for use in the aluminum industry. A related system is adapted to remove metal content from coke and convert fuel grade coke to anode grade coke.

Abstract: Generally, the present disclosure is directed to systems and methods of quantizing a database with respect to a novel loss or quantization error function which applies a weight to an error measurement of quantized elements respectively corresponding to the datapoints in the database. The weight is determined based on the magnitude of an inner product between the respective datapoints and a query compared therewith. In contrast to previous work, embodiments of the proposed loss function are responsive to the expected magnitude of an inner product between the respective datapoints and a query compared therewith and can prioritize error reduction for higher-ranked pairings of the query and the datapoints. Thus, the systems and methods of the present disclosure provide solutions to some of the problems with traditional quantization approaches, which regard all error as equally impactful.

Abstract: The present invention relates to a process for production of anisotropic coke from a hydrocarbon feedstock and a system for producing the same. More particularly, the present invention relates to a thermal cracking of heavy petroleum residue producing petroleum coke and lighter hydrocarbon products. The invented process utilizes a novel scheme for production of a premium quality coke from primarily, a clarified oil feedstock. Clarified oil from fluid catalytic cracking unit is routed through a process scheme comprising a separator column, hydrotreatment section and an aromatic extraction section to create an ad-mix of effluents which form the feedstock to a thermal cracking unit. Premium quality anisotropic coke is produced in the thermal cracker reactor drums under tailor made process conditions employing the said feedstock.

Abstract: A sensor data forecasting system for urban environment using deep learning model is provided.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for privacy preserving training of a machine learning model.

Abstract: Embodiments of the invention relate to a membrane-electrode assembly comprising a membrane structure with an anode layer, a cathode layer, and a membrane layer, wherein the membrane layer is positioned between the anode layer and the cathode layer. The membrane-electrode assembly furthermore comprises an anode-side gas-diffusion layer arranged on the anode layer and a cathode-side gas-diffusion layer arranged on the cathode layer. Furthermore, at least one of the anode-side gas-diffusion and the cathode-side gas-diffusion layers has a structure on the side facing away from the membrane structure. According to some embodiments, the structure comprises a plurality of columns for forming a laterally-open flow field, wherein the columns have support surfaces for supporting a bipolar plate.

Abstract: A distribution unit for distributing a multi-phase fluid mixture is disclosed. The distribution unit includes a distribution body defining a first passage, and a first distal body portion having a plurality of first slots. The distribution body includes a second distal body portion having a plurality of second slots distributed on a side wall of the second distal body portion. Each of the plurality of second slots is adapted to accommodate a baffle plate. The second distal body portion includes at least one aperture formed on a bottom wall of the second distal body portion. The plurality of first slots, the plurality of second slots, and the at least one aperture are in fluid communication with the first passage to discharge the flow of the multi-phase fluid.

Abstract: Systems and methods for low bias negative sampling of classes according to the sampled softmax method are described herein. The systems and methods can include training a machine-learned model for classifying inputs into one or more classes of a plurality of classes, each of the plurality of classes having an associated class embedding in a plurality of class embeddings. The systems and methods can include selecting, by the one or more computing devices, one or more negative classes from the plurality of classes based at least in part on a probability distribution approximating a softmax distribution, wherein the probability distribution is determined based at least in part on a Random Fourier Features map.

Abstract: The present invention provides a catalyst composition comprising rare earth exchanged USY zeolite (REUSY); pentasil zeolite; phosphorous compound; clay, silica, alumina, and spinel to enhance the catalytic activity and selectivity for light olefins in FCC operation conditions. The present invention also provides a process for the preparation of Light olefin enhancing catalyst composition with high propylene yield and coke selectivity.

Abstract: Generally, the present disclosure is directed to systems and methods that perform adaptive optimization with improved convergence properties. The adaptive optimization techniques described herein are useful in various optimization scenarios, including, for example, training a machine-learned model such as, for example, a neural network. In particular, according to one aspect of the present disclosure, a system implementing the adaptive optimization technique can, over a plurality of iterations, employ an adaptive effective learning rate while also ensuring that the effective learning rate is non-increasing.

Abstract: Systems and methods for assessing vehicle operation are provided. According to certain aspects, an electronic device may receive and analyze image data depicting an individual located within a vehicle. The electronic device may also access and compile certain data related to an environment of the vehicle, and specifically to the vehicle's position relative to the sun. According to embodiments, the electronic device may determine, based on the accessed data, whether the vehicle is traveling into a direction of the sun, and may generate and record an indication of the same.